[ ]productinterview
Run a free mock

product sense · standard

Design an alarm clock for the blind

Design an alarm clock for people who are blind.

Updated Jun 2026 Calibrated to the strong-hire bar

The classic failure on this prompt is treating it as an accessibility feature checklist: add voice control, add braille labels, make the alarm louder. Every candidate does this. Microsoft interviewers, who documented this question on Glassdoor including a harder senior-loop variant (“design for someone who is deaf and blind at the same time”), are not scoring feature quantity. They are scoring whether you can identify the most underserved sub-segment, articulate why existing products fail that segment specifically, and make a prioritization call you can defend.

Scope it before you segment

Before segmenting, ask: “Are we designing a standalone physical clock, or is the brief really about waking up reliably and independently?” The JTBD is not “tell time.” It is: wake up at the intended moment without anxiety that the alarm was set wrong or will be missed.

That reframe matters because the phone ecosystem already solves most of this for most blind users. VoiceOver and TalkBack let any iPhone or Android user set and confirm alarms entirely by voice, for free. Existing talking clocks (Cirbic, BESTYASH, $20-40 on Amazon) cover the auditory baseline. A standalone clock must justify itself against that. Naming this earns credit before you segment anyone.

Segments that actually differ

Most candidates split “fully blind versus partially blind” and immediately collapse to fully blind with no real argument. The three segments that carry meaningfully different design constraints:

Congenitally blind adults. Expert AT users; VoiceOver, TalkBack, and JAWS are native to them. Their real unmet pain is alarm-set confirmation anxiety. The $20 talking clock covers the baseline but gives no confirmation feedback.

Late-onset blind adults. Have sighted muscle memory and find software AT steep. A dedicated physical device has real value the phone ecosystem does not provide.

Deaf-blind adults. Neither audio nor visual channels work; haptic and braille are the only modalities. Estimated 70,000 to 100,000 in the US (Helen Keller National Center). The phone ecosystem is entirely audio-first and offers nothing for this group. The Dot Watch uses a refreshable braille display on the wrist at around $290, but its alarm-setting UX is poor. No mass-market product reliably reaches them.

Choose deaf-blind as the focus: highest severity of unmet need; existing products entirely fail this segment; no competition from the phone ecosystem; and Microsoft has institutional credibility in accessibility (Xbox Adaptive Controller, Seeing AI) that makes this a plausible strategic surface.

Strong answer

strong

"Before I segment, I want to name a prior question: are we committed to a standalone physical clock, or is the brief about waking up reliably and independently? VoiceOver and TalkBack already solve this for most blind users on their phones. If the brief is open, I'd argue software first. But if it's a physical device, I'd focus on deaf-blind adults: they're the one segment where the phone ecosystem fails entirely because it's audio-first, and audio isn't a channel for them. Roughly 70,000 to 100,000 people in the US, zero mass-market product that reliably reaches them.

Their core pain is confirmation: no feedback channel exists to verify the alarm was set. Secondary pain: phone vibration on a nightstand is insufficient without audio backup.

The product: a wrist-worn haptic band with a single refreshable braille cell on the inner wrist. No screen. One squeeze-to-confirm gesture. Alarm setting via a paired app with VoiceOver support for users with a caretaker, or voice relay for users with residual hearing. The band outputs braille confirmation within three seconds: 'ALM 7:00 SET.' Alarm delivery: escalating wrist vibration plus an optional bed-shaker output jack. On-device AI (running offline, no cloud dependency) learns the routine and pushes proactive braille output before sleep: 'No Thursday alarm set, same as Wednesday?' Zero interaction required on a routine morning.

Viability: consumer TAM is constrained because the phone works for most users. The real channel is institutional: hospitals, elder care, and schools for the blind where occupational therapists make procurement decisions and the product competes on clinical outcomes. Microsoft could white-label to hearing and vision device makers already serving deaf-blind overlap.

Metrics: alarm-set confirmation rate within three seconds, target 99%; false-negative rate (alarm fired, user did not wake), target below 2%; time-to-set for a new user, target under 45 seconds; 90-day retention versus reverting to caretaker, target 70%; NPS from the OT procurement channel."

Weak answer

weak

"I'd add voice control, braille labels on the buttons, and a louder alarm sound." This fails in compounding ways. Voice-controlled alarm setting already exists free on every smartphone: the candidate is proposing to rebuild iOS accessibility in clock hardware. Braille literacy among blind Americans is roughly 10%, so anchoring on it as the primary modality signals no research. A louder alarm does nothing for deaf-blind users, who represent the hardest constraint in this prompt. No segment, no viability argument, no measurable metric. The interviewer reads: memorized feature list, zero ecosystem awareness.

The 2026 angle

Voice control, on-device NLP, and haptic hardware are commodity. A talking clock costs $25 on Amazon. The question is what “genuinely usable” means for the most underserved sub-segment.

For deaf-blind users, the entire audio-first solution space collapses and the design problem becomes purely tactile. A candidate who names deaf-blind as the focus, explains why existing products fail them, and proposes a haptic-plus-braille solution with an institutional go-to-market is operating at a different level from the candidate running a feature checklist from 2018 accessibility design blogs.

The lovable bar: the device knows your schedule, confirms proactively, and requires zero interaction on a routine morning. The viable bar: sold through occupational therapists and institutional procurement, not competing on Amazon against a phone that already works for most blind users. The deaf-blind US sub-segment is 70,000 to 100,000 people: too small for consumer retail, exactly right for an institutional channel with a clear clinical outcome.

Framework

Asked at

Related